A good review system will have some abstraction from the final score. The tendency for a 1-10 score to be polarized higher than it should be means that you're really not getting a good picture of actual performance. There's too much bias in just choosing a final, concrete score.
Here's the solution:In short, Itemized service reviews and aggregated total scores.
For each of the services rendered, the provider gets a 1-10 score.
This means that, if the provider performs DFK, BBBJ, CFS, they would get 3, 1-10 scores for their work.
Then, the scores are averaged to determine the aggregate performance score.
-Performing more services would allow a provider to pad their scores with additional services only if the services provided were actually worthy of high scores.
-Performing a limited number of services would allow a provider to score higher only if said services were performed well, but would penalize a provider more if any of those services were performed poorly. It does not explicitly penalize providers for not performing certain services.
-Additional data capture would allow the search function to be more granular. Clients would be able to search for who was performing well at a particular service.
-Reviews would be more useful to providers because they would have granular feedback on which services they offer score highly and which services they could work on.
In addition to the current rating system, TER should have 1 question: please rate your overall satisfaction with the experience on a scale of 1-10. Consider the following factors:
* Value
* Desire to exceed your expectations
* Quality of experience
* Screening and booking process
* Range of services offered
And perhaps 4-8 other factors. Individuals should determine the trade space between factors and/or the relative importance of each factor. That should augment the existing rating scheme. Scores should be rolled up for agencies.
... when deciding what Performance rating to give on a review.
A good review system will have some abstraction from the final score. The tendency for a 1-10 score to be polarized higher than it should be means that you're really not getting a good picture of actual performance. There's too much bias in just choosing a final, concrete score.
Here's the solution:
In short, Itemized service reviews and aggregated total scores.
For each of the services rendered, the provider gets a 1-10 score.
This means that, if the provider performs DFK, BBBJ, CFS, they would get 3, 1-10 scores for their work.
Then, the scores are averaged to determine the aggregate performance score.
-Performing more services would allow a provider to pad their scores with additional services only if the services provided were actually worthy of high scores.
-Performing a limited number of services would allow a provider to score higher only if said services were performed well, but would penalize a provider more if any of those services were performed poorly. It does not explicitly penalize providers for not performing certain services.
-Additional data capture would allow the search function to be more granular. Clients would be able to search for who was performing well at a particular service.
-Reviews would be more useful to providers because they would have granular feedback on which services they offer score highly and which services they could work on.